论文部分内容阅读
图像光流的计算不需要在图像序列中建立特征之间的对应关系 ,因此光流法在计算机视觉的众多领域 ,包括运动物体的参数估计和目标跟踪方面都有广泛的应用。由于红外图像的噪声相对较大 ,光流法很少用于红外图像中目标的运动参数估计和跟踪。这里 ,使用几种常用的光流计算方法对部分实际红外图像进行了光流场计算。结果表明 ,当选择合适的方法或对计算方法进行一定的改进时 ,这些红外图像可以得到比较接近实际情况的目标光流场 ,进而应用于红外图像中的目标分割、运动状态分析与目标跟踪等领域。
Therefore, the optical flow method is widely used in many fields of computer vision, including the parameter estimation and target tracking of moving objects. Due to the relatively large noise of infrared images, the optical flow method is rarely used for the estimation and tracking of the motion parameters of the target in infrared images. Here, several common optical flow calculation methods are used to calculate the optical flow field of some actual infrared images. The results show that when selecting the appropriate method or improving the calculation method, these infrared images can get the target optical flow field which is close to the actual situation, and then applied to the target segmentation, motion state analysis and target tracking in infrared images field.